Quote:
Originally Posted by Ian Robinson
Would be a great honours year thesis project for a electrical engineering or computer engineering or mechatronics engineering student .... I might suggest it to Newcastle Uni.
|
This may be a good idea..
And, if the paper is made available to some Chinese company
, excellent!
Because, the cost of hardware for project like this is not that much... $200 max for prototyping, including controller evaluation boards. In my particular case, since boards are already available, I have to buy only some parts, and of course, mechanics...
Still way better than PC.
BTW, TvGuider has its website, here:
http://foto.astronomy.cz/TVGuider.htm
On system properties, you are right, it is not linear, because of the mechanical backlash, for the start.
Then, the gain will be variable because of PE...
But I am sure it can be treated as two independent 1-dimensional loops.
The challenge will be to design the response such that the stability is adequate for variety of mounts (some have steppers, some have DC motors coupled in their own control loops).. or perhaps not because the response of those loops will be relatively fast compared to to our guiding loop, so maybe it can be ignored for the start.
I think I'd better have a good look at Niemi's souce code :-)